To have or not Byte Order Mark (BOM) in UTF-8 text files?(Linux)

Is it advisable to have or not Byte Order Mark (BOM) in **UTF-8 text** files on Linux? Is it correct to say byte order (even for multi-byte characters) is already strictly defined/fixed in UTF-8 standard? Would you say Linux soft & CLI utilities are generally more compatible with UTF-8 (w/o BOM) or with UTF-8 BOM? Any examples of incompatibility? Example soft I primary use: Doublecmd, terminal, Libreoffice, bash (scripts),Kate. I'm doing encoding conversion of many non-ASCII text files to **UTF-8**, so need to consciously decide whether to choose UTF-8 or UTF-8 BOM. Text files include: HTML, Bash scripts, C/C++ code, TeX etc. Thank you!